Although antiretroviral therapy (ART) has transformed HIV infection into a chronic disease, it is not curative. ART does not target latent, integrated provirus harbored in memory CD4+ T cells. These cells are relatively rare, possibly on the order of 1 in a million frequency, and it is not yet possible to identify all of them prospectively. Any method that aims to eliminate or inactivate provirus must be highly efficient and work in vivo. It is also known that disrupting ccr5 can result in resistance to HIV and is the basis of the single individual cured of HIV. A technique that could efficiently target and disrupt ccr5 in CD4+ T cells and other cell types in vivo would be preferable over current methodologies that rely on ex vivo manipulation. For this application we propose to develop and optimize a helper-dependent adenoviral (HDAd) vector termed 5/35++ that is capable of targeting and efficiently transducing CD4+ T cells and hematopoietic progenitors. HDAds are essentially ?gutted? adenoviruses that can package up to 30 kb of foreign DNA and also rely on a helper virus (HV) for amplification. In addition, as opposed to adeno-associated vectors, once the HDAd has been amplified it is facile to re-amplify the HDAd to extremely high titers of ~1013 vp/ml. Although most HVs used for amplification are serotype 5-based, it was recently reported that an Ad5/F35 HV (5/35++) has enhanced in vivo targeting for hematopoietic stem/progenitor cell gene delivery. We propose using this novel Ad5/F35 HV to target HDAds to human CD4+ T cells and other hematopoietic progenitors, first in vitro and then in vivo, the latter in humanized mouse models in collaboration with my colleague Dr. Priti Kumar at Yale. Initially the HDAds will encode eGFP and luciferase to allow in vitro FACS analysis and in vivo imaging, respectively, but other constructs will include the components of multiple, orthogonal CRISPR systems (both nuclease and guide RNAs) for more efficient targeting of host gene ccr5 in vivo. This includes encoding multiple guide RNAs, each targeting ccr5 and separated by distinct tRNAs, in RNAPIII promoter-driven constructs. These will be tested both in vitro and in vivo, the latter in an HIV challenge model in humanized mice. In addition, in order to expand the repertoire of HVs capable of targeting T cells and progenitors, HVs based on several other adenovirus serotypes with those properties will be constructed, amplified, and tested in vitro. This should allow repeated injections of HDAds in immunocompetent animals and man, without the concern for the development of cross-reactive neutralizing antibodies that would limit efficiency. At the end of the 24 month funding period it is hoped that the potential of this novel HV/HDAd system for targeting both CD4+ T cells and hematopoietic progenitors in vivo will have been realized. Although the application focuses on targeting ccr5, similar methods may be applied for inactivating integrated HIV provirus, especially given the fact that adenovirus can be retargeted to other cells using fiber gene fusions and given the recent result that CD32a may be a biomarker for latently HIV-infected memory T cells.